A group block distribution strategy for a heterogeneous machine

Abstract

This paper discusses the data distribution problem for inherently sequential algorithms, such as the LU factorization in linear algebra, when computed on heterogeneous machines. These algorithms present additional difficulties to optimize the processing time due to the fact that the computational load for data matrix columns increases with their index, requiring a fine tuned load assignment and distribution. A new static load distribution strategy is presented that achieves better results than other algorithms developed for the same purpose as shown in the results section. The strategy is called static because it distinguishes the computation environment implementation from the algorithm implementation, as explained below.